Changes in collagen fibril network organization and proteoglycan distribution in equine articular cartilage during maturation and growth

Abstract

The aim of this study was to record growth-related changes in collagen network organization and proteoglycan distribution in intermittently peak-loaded and continuously lower-level-loaded articular cartilage. Cartilage from the proximal phalangeal bone of the equine metacarpophalangeal joint at birth, at 5, 11 and 18 months, and at 6-10 years of age was collected from two sites. Site 1, at the joint margin, is unloaded at slow gaits but is subjected to high-intensity loading during athletic activity; site 2 is a continuously but less intensively loaded site in the centre of the joint. The degree of collagen parallelism was determined with quantitative polarized light microscopy and the parallelism index for collagen fibrils was computed from the cartilage surface to the osteochondral junction. Concurrent changes in the proteoglycan distribution were quantified with digital densitometry. We found that the parallelism index increased significantly with age (up to 90%). At birth, site 2 exhibited a more organized collagen network than site 1. In adult horses this situation was reversed. The superficial and intermediate zones exhibited the greatest reorganization of collagen. Site 1 had a higher proteoglycan content than site 2 at birth but here too the situation was reversed in adult horses. We conclude that large changes in joint loading during growth and maturation in the period from birth to adulthood profoundly affect the architecture of the collagen network in equine cartilage. In addition, the distribution and content of proteoglycans are modified significantly by altered joint use. Intermittent peak-loading with shear seems to induce higher collagen parallelism and a lower proteoglycan content in cartilage than more constant weight-bearing. Therefore, we hypothesize that the formation of mature articular cartilage with a highly parallel collagen network and relatively low proteoglycan content in the peak-loaded area of a joint is needed to withstand intermittent stress and shear, whereas a constantly weight-bearing joint area benefits from lower collagen parallelism and a higher proteoglycan content.

Fig. 1

(A) Study design and time points for quantitative analysis of equine articular cartilage collagen network parallelism and proteoglycan (PG) content in the equine metacarpophalangeal (MCP) joint. PLM, polarized light microscopy of collagen; SAF, digital densitometry of the PG distribution after safranin-O staining. The age range covers newborn (day 0), growing and maturing (5, 11 and 18 months) and adult (6–10 years) animals. n= number of animals. (B) Sampling sites of cylindrical osteochondral plugs for microscopy. Site 1 is situated at the dorsal edge and site 2 in the mid-region (central fovea) of the proximal articular surface of the proximal phalangeal bone.

Fig. 3

The average degree of collagen parallelism (mean ± SD) through the cartilage thickness, from the surface to the osteochondral junction, measured with quantitative polarized light microscopy. 0% indicates random course of fibrils [parallelism index (PI) = 0] and 100% indicates perfect parallelism (PI = 100). The cartilage thickness is expressed using a relative scale (10–100%, x-axis) from the cartilage surface to the osteochondral junction. Newborn (day 0), growing and maturing (5, 11 and 18 months), and adult (6–10 years) horses (A–E). *P < 0.05, **P < 0.01, ***P < 0.001 after Bonferroni adjustment for comparison of the individual fractions between sites 1 and 2. The P-values in (A–E) indicate the significance of the overall difference between the parallelism distributions in sites 1 and 2. Wilcoxon’s signed ranks test for paired observations. NS, not significant.

Fig. 2

Typical example of polarized light microscopy slide from site 1, demonstrating the degree of parallelism of the fibrils in the collagen network at day 0 (A), at 18 months (B) and in the adult animal (8 years) (C). Bright intensity indicates high and dark intensity represents low parallelism of collagen fibrils. The decrease of the dark area indicates that there is increased parallelism in the more superficial cartilage layers over time. The asterisks show the interface between articular cartilage and bone. Scale bar, 500 μm.

Fig. 4

Safranin-O stain absorbance (mean ± SD) as a measure of zonal proteoglycan content and distribution at day 0 (A), at 18 months (B) and in adult animals (6–10 years) (C). Fractions of articular cartilage start from the surface and end at the osteochondral interface. The four fractions correspond roughly to the superficial, intermediate, deep and calcified zones of cartilage. AIOD, area-integrated optical density of safranin-O. Wilcoxon’s signed ranks test for paired observations. NS, not significant.